Quantitative Study Of Lung Structure In COPD Based On Low-dose Karl Iterative Reconstruction

ObjectiveBased on the low-dose Karl iterative reconstruction technique,the quantitative study of the lung structure of COPD and its clinical application value were discussed.This research contains three parts.1.The first part:Based on the conventional dose and low-dose Karl iterative reconstruction technique,the effects of different levels of iterative reconstruction techniques on the chest image quality of chronic obstructive pulmonary disease?COPD?CT were explored,and the optimal iteration level was found.2.The second part:under the premise of iterative advantage,using the"FACT-Digital lung ^TM"to analyze the difference between the conventional index and the low-dose conventional reconstruction FBP and the low-dose Karl in the quantitative analysis of the lobulation and bronchial grading.Under the premise of CT low-dose scanning,whether different Karl iterative reconstruction algorithms have an effect on CAD quantitative indexes,and then analyze the clinical value of low-dose iterative reconstruction technique in CT quantification of COPD patients.3.The third part:Based on the conclusions of the second part,the conventional double-gas phase?inhalation and expiration?images of low-dose COPD patients were reconstructed with FBP and different Karl iterative reconstruction levels.Correlation analysis between indicators and lung function was conducted to investigate the effect of Karl iteration technique on the correlation of CT lung function and to analyze the value of this technique in the diagnosis of COPD.Part one1.Materials and methodsThe dual-gas chest CT data of patients with chronic obstructive pulmonary disease from May 2018 to December 2018 in our hospital were initially collected.All the scanned persons signed informed consent.Inclusion criteria:Complies with the diagnostic criteria for?COPD in the Global Strategy for the Diagnosis,Treatment and Prevention of GOLD Chronic Obstructive Pulmonary Disease in 2018?:?1?There are no other diseases that cause changes in lung volume,such as thoracic deformities;?2?Inhaled bronchodilator FEV1/FVC<0.70;?3?The CT chest double weather image is clear and complete,and the lung tissue is clearly displayed.Exclusion criteria:?1?Poor patient coordination,the image is unclear;?2?Lesions affecting lung volume,such as history of intrapulmonary surgery?after lobectomy or segmentectomy?,chest deformities,etc;?3?Intrapulmonary tumor or tumor-like leision;?4?Diseases affecting CT images,such as most infections of the lungs,severe atelectasis,and lung consolidation;?5?Other serious diseases such as heart,liver,kidney and other diseases.Conventional dose?120KV,150mAs?and low-dose?120KV,80mAs?CT scan images were reconstructed by traditional filtered back projection?FBP?and 1-9 grade Karl iterative reconstruction.The reconstructed image is transmitted to the post-processing station for subjective and objective index evaluation.The data is analyzed by SPSS20.0statistical analysis software,and the measurement data is expressed by ???±s;Low-dose CT scan images were performed on FBP and Karl iterative images reconstructed at 1-9 level.The objective indexes of aortic CT,aortic SD,SNR and CNR were analyzed.P<0.05 was statistically significant.CTDIVol,DLP and ED of conventional dose and low-dose CT scan were used in paired sample t-test.P<0.05 was statistically significant.The consistency of the image quality scores of the two senior physicians was tested by Kappa test.Kappa?0.75 was consistent,0.4?Kappa<0.75 was moderate,Kappa<0.4 was poor consistency;Subjective image quality scores were analyzed by analysis of variance.P<0.05 was statistically significant.The highest iterations of conventional doses of FBP and low-dose subjective scores were analyzed by paired-sample t-test.P<0.05 was statistically significant.2.ResultsFinally,30 patients with chronic obstructive pulmonary disease were collected,including25 males and 5 females.1.The SD value,SNR value and CNR value of the low-dose?120KV,80mAs?CT image FBP and the 1-9 grade Karl iterative reconstruction were statistically significant?P<0.05?.The CT values of the images in each group were not statistical.Significance?P=1>0.05?.The SD values of the 1-9-level Karl iterative reconstruction images are lower than FBP.With the increase of the Karl iteration level,the SD value increases gradually.The SNR value and CNR value of the 1-9Karl iterative reconstruction image are higher than FBP,and at the Karl iteration level.At the same time,the SNR value and CNR value increase gradually;the CT value and FBP of the 1-9Karl iterative reconstruction image change little,and with the increase of the Karl iteration level,the CT value has no obvious change.2.The conventional doses?120KV,150mAs?and low dose?120KV,80mAs?radiation doses CTDIVol,DLP and ED were statistically significant?P<0.05?,low doses of radiation dose ED and conventional doses decreased significantly?About 47%?.3.The difference of image quality subjective scores between low-dose CT images FBP and 1-9 grade Karl iterative reconstruction was statistically significant?P<0.05?.The subjective scores of the images after different levels of Karl iterative reconstruction are higher than FBP,and the highest subjective score is Karl iteration 5.4.There was no significant difference in CT,SD,SNR and CNR between the conventional dose FBP and the low dose Karl iteration 5?P>0.05?.The changes of CT value,SD value,SNR value and CNR value of the two groups were small.3.summaryUnder low-dose?120KV,80mAs?scanning conditions,the effective dose is reduced by about 47%while maintaining image quality.The image iterative reconstruction technique has significantly improved image noise reduction performance compared with conventional reconstruction FBP,and the image SNR and CNR also have As the level of Karl iteration continues to increase,the image's noise reduction capability also increases,and the Karl5 score is highest in subjective evaluation,Low dose Karl 5 can be applied to CT scans of patients with chest COPD.Part two1.Materials and methodsInitially,33 patients with chronic obstructive pulmonary disease who were admitted to our hospital from May 2018 to December 2018 were enrolled.All the scanned persons signed informed consent.The inclusion criteria and exclusion criteria are consistent with the first part.Conventional dose?120KV,150mAs?and low-dose?120KV,80mAs?CT scan images were reconstructed by traditional filtered back projection?FBP?and 1-9 grade Karl iterative reconstruction.All images were imported into the"FACT-Digital lung^TM"for lobectomy and quantitative bronchoscopy.The data were analyzed by SPSS20.0 statistical analysis software,and the measurement data were expressed by ???±s;the data obtained by conventional dose FBP and low-dose FBP images were analyzed by paired sample t test;the data obtained by low-dose conventional FBP and low-dose 1-9-level Karl iterative reconstruction images were performed.One-way analysis of variance,all statistical analysis p<0.05 considered that the difference was statistically significant;when there were differences,LSD was used for pairwise comparison,p<0.05 was considered statistically significant.2.Results1.There was no significant difference in LV,LAA910%,LAA950%,EI and mean density?MLD?between conventional dose FBP and low dose FBP and different Karl iterative grades of whole lung and lung parenchyma?p>0.05?;conventional dose The total lung volume,LAA910%,LAA950%,EI measured by FBP were lower than that of low dose FBP and different Karl iterative reconstruction.The MLD measured by conventional dose FBP was higher than that of low dose FBP and different Karl iterative reconstruction.2.Conventional dose FBP and low-dose FBP and different iterative grades of Pulmonary Pulmonary Fissure,left lung oblique surface area and left lung oblique fracture integrity,right lung oblique surface area and right lung oblique fracture integrity,right lung level There was no significant difference in the fissure surface area and the horizontal integrity of the right lung?p>0.05?.3.There was no significant difference in the number of bronchus,length and bronchial volume between the conventional dose FBP and the low-dose FBP and the different lung bronchus groups?p>0.05?.3.summaryThere were no significant changes in the quantitative analysis of lung parenchyma and bronchial CAD between conventional and low-dose reconstituted images;LAA 950%and EI were more sensitive to CAD quantification in lung parenchyma quantitative indicators;Low-dose iterative reconstruction can be applied to clinical quantitative studies,and LAA 950%and EI quantitative indicators have more clinical significance.Part three1.Materials and methodsInitially,33 patients with chronic obstructive pulmonary disease who were admitted to our hospital from May 2018 to December 2018 were enrolled.All the scanned persons signed informed consent.The inclusion criteria and exclusion criteria are consistent with the first part.The images obtained by low-dose?120KV,80mAs?CT scan were reconstructed by traditional filtered back projection?FBP?and 1-9 grade Karl iterative reconstruction.All images were imported into the"FACT-Digital lung ^TM"for quantitative measurements.A routine lung function test?PET?was performed on all subjects.The data were analyzed by SPSS20.0 statistical analysis software,and the measurement data were expressed by ???±s.The Shapiro-Wilk test?W test?was used for normality test.According to the test results,the correlation between CT quantitative indicators and lung function was analyzed by Pearson correlation analysis.Spearman rank correlation analysis,the above statistical analysis were statistically significant at P<0.05;correlation coefficient|r|was uncorrelated at 0.0-0.2,weak correlation at 0.2-0.4,moderate correlation at 0.4-0.6,0.6-0.8 is strongly correlated,and 0.8-1.0 is strongly correlated.2.Results1.Low-dose conventional reconstruction FBP and 1-9 grade Karl iterative reconstruction images The dual-gas total lung volume difference?Vin-Vex?was strongly correlated with FEV1%,FEV1/FVC%?r values were 0.60,0.62,respectively;P=0.01?;Vin-Vex was moderately correlated with MEF50 and MEF75?r values were 0.51,p<0.05?;Vin-Vex was not associated with FVC?r value was 0.22,P>0.05?;total lung volume difference?Vin-Vex?and There was no correlation between PEF?r value was 0.16,P>0.05?;as the iterative level increased,the correlation of lung function related indicators did not change.2.There was no correlation between low-dose conventional reconstructed FBP and 1-9Karl Iterative Reconstructed Image Biphasic Total Lung Volume Ratio?Vin/Vex?and FEV1%,FEV1/FVC%,PEF,MEF50 and MEF75?P>0.05?.3.Low-dose conventional reconstruction FBP and 1-9 grade Karl iterative reconstruction image The percentage of total gas volume change in the dual-gas phase??Vin-Vex?/Vin?has no correlation with FEV1%,FEV1/FVC%,PEF,MEF50 and MEF75?P>0.05?.4.Low-dose conventional reconstruction FBP and 1-9 grade Karl iterative reconstruction images Dual-gas phase density difference?Din-Dex?was strongly correlated with FEV1%,FEV1/FVC%,MEF75?r values were-0.60,-0.72,-0.63,respectively;P The values were 0.01,0.00,and 0.01?;Din-Dex was moderately correlated with PEF and MEF50?r values were-0.54,-0.56,and p<0.05,respectively?;Din-Dex was not associated with FVC?r value was-0.20,P>0.05?.As the level of iteration increased,Din-Dex increased slightly,and the correlation with lung function related indicators did not change.5.Low-dose conventional reconstruction FBP and 1-9 grade Karl iterative reconstruction images The dual-gas whole lung density ratio?Dex/Dex?is strongly correlated with FEV1%,FEV1/FVC%,MEF75?r values are 0.60,0.71,0.62,respectively;P value?0.01,0.00,0.01);Dex/Dex was moderately correlated with PEF MEF50?r values were 0.53,0.55,p<0.05?;Dex/Dex was not associated with FVC?r value was 0.20,P>0.05?;As the level of iteration increased,the overall lung density ratio?Vex/Dex?and its correlation with lung function related indicators did not change.6.The low-dose conventional reconstruction FBP and the 1-9 grade Karl iterative reconstruction image showed that the percentage of total lung density change?Din-Dex?/Din was strongly correlated with FEV1%,FEV1/FVC%and MEF75?r values were 0.60,respectively?.0.71,0.62;P values were 0.01,0.00,0.01);??Din-Dex?/Din?was moderately correlated with PEF and MEF50?r values were 0.53,0.55,p<0.05,respectively?;??Din-Dex?/Din?was not associated with FVC?r=0.20,P>0.05?;percentage change in total lung density??Din-Dex?/Din?and its function with lung function as iterative level increased The correlation of related indicators has not changed.7.Low-dose conventional reconstruction FBP and 1-9 grade Karl iterative reconstruction images inhalation gas whole pulmonary emphysema index?EI?was strongly correlated with FEV1%,FEV1/FVC%,MEF75?r values were-0.60,-0.71,-0.62,respectively;P The values were 0.01,0.00,and 0.01?;EI was moderately correlated with PEF and MEF50?r values were-0.53,-0.55,p<0.05,respectively?;EI was not associated with FVC?r value was-0.20,P>0.05?;The increase in the iterative level did not change the overall pulmonary emphysema index?EI?and its association with lung function-related indicators.3.summaryUnder low dose?120KV,80mAs?scanning conditions:?1?CT quantitative methods can quantitatively analyze and evaluate the pulmonary function of COPD,and can also assess the respiratory function of patients with COPD,which can play a role in assisting diagnosis to a certain extent.?2?The quantitative correlation between the three methods and lung function was found to be strongly correlated with FEV1%,FEV1/FVC%in lung function indicators.?3?Din/Dex is the best indicator for evaluating air retention in lung density measurement,but this method has certain defects.However,this study found that the difference of double gas phase density and the change of dual gas phase density also have a good correlation with lung function index.Therefore,the correlation with lung function can be analyzed as a supplement to a certain extent,thereby increasing the reliability of the study.Conclusions?1?Karl iterative reconstruction can significantly reduce the noise of the image and improve the image SNR and CNR,and the effect becomes more obvious as the iteration increases.At the same time as the dose is reduced?approximately 47%reduction?,the image after Karl iteration can be applied to the clinic,and the optimal Karl iteration level is 5,it can be applied to CT scans of patients with chest COPD.?2?Under low-dose?120KV,80mAs?scanning conditions,Karl iterative reconstruction can be used for quantitative analysis of lung parenchyma and bronchial CAD,especially the application of LAA950%and EI%,which can better improve the clinical application of various aspects of COPD.?3?Under low-dose?120KV,80mAs?scanning conditions,the method used for quantitative CT measurement can play an important clinical role in the early evaluation of COPD to a certain extent,and can improve the accuracy and even replace the lung function test under certain conditions.Part of the role of lung function,thereby improving the clinical application prospects of early diagnosis,treatment and prognosis of COPD.